ppyoloe 训练损失函数为nan

[sensenets123456]

问题确认 Search before asking

• [X] 我已经查询历史issue，没有发现相似的bug。I have searched the issues and found no similar bug report.

Bug组件 Bug Component

Training

Bug描述 Describe the Bug

ppdet\modeling\heads\ppyoloe_head.py def _bbox_loss(self, pred_dist, pred_bboxes, anchor_points, assigned_labels,assigned_bboxes, assigned_scores, assigned_scores_sum): 函数中的代码loss_cls /= assigned_scores_sum 当assigned_scores_sum为0时候，得到loss_cls 的值为inf

复现环境 Environment

linux

Bug描述确认 Bug description confirmation

• [X] 我确认已经提供了Bug复现步骤、代码改动说明、以及环境信息，确认问题是可以复现的。I confirm that the bug replication steps, code change instructions, and environment information have been provided, and the problem can be reproduced.

是否愿意提交PR？ Are you willing to submit a PR?

• [X] 我愿意提交PR！I'd like to help by submitting a PR!

[nemonameless]

请再详细描述下Bug复现步骤、代码改动说明、以及环境信息，谢谢。尤其是环境，几卡总bs的改动。 assigned_scores_sum几乎不大可能为0。如果还不行可以手动设置为1。

[sensenets123456]

环境：ubantu，从官网拉下的paddle镜像：paddledetection-cuda11.2-cudnn8:v1.0， 2卡，batch size设置为1，图片大小为1920*1920，在citr数据集上训练，